Manipulative Attacks and Group Identification

TL;DR

This paper surveys manipulative attacks on group identification processes, analyzing their computational complexity and introducing new results on control problems in various profile settings.

Contribution

It provides a comprehensive overview of manipulative attacks in group identification and presents new complexity results for control problems in binary, ternary, and r-profiles.

Findings

Most results are aggregated from existing literature.

New complexity results for control in binary, ternary, and r-profiles.

Parameterized complexity analyses for many problems.

Abstract

The group identification problem asks to identify a socially qualified subgroup among a group of individuals based on their pairwise valuations. There are several different rules that can be used to determine the social qualification status. In this work, we consider the consent rules, the consensus-start-respecting rule, and the liberal-start-respecting rule. In the context of group identification, a manipulative attack is the attempt by an outsider to influence the outcome of the selection process through certain means of manipulation. These means include adding, removing, or partitioning individuals, as well as bribing individuals to change their opinion. In this work, we provide an overview of manipulative attacks in group identification as well as group identification with partial profiles. In particular, we study the computational complexity of the corresponding problems. Most results presented in this work are aggregated from the literature, but we also show results for previously unstudied problems; these include general and exact group control in binary profiles and in ternary profiles, as well as constructive group control in $r$-profiles. For many considered problems, we also study the parameterized complexity.